Pruning cost complexity and pruning time complexity. This set of artificial intelligence multiple-choice questions %26 Answers (MCQ) focuses on “decision trees”. Pre-pruning: The tree is pruned by stopping its construction ahead of time. Post-pruning: This approach removes a subtree from a fully developed tree.
Cost complexity is measured by the following two parameters -. The process of adjusting the decision tree to minimize the “misclassification error” is called debugging. It is of 2 types: pre-pruning and subsequent pruning. Once the tree series has been created, the best tree is chosen using generalized precision measured by a training set or cross-validation.
So, to solve those problems, we first created the decision tree and then used the error rates to properly prune the trees. Pruning is a data compression technique in machine learning and search algorithms that reduces the size of decision trees by eliminating sections of the tree that are not critical and are redundant for classifying instances. In contrast, other tree algorithms, such as ID3, can produce decision trees with nodes that have more than two children.